Diffusion probabilistic models (DPMs) have been shown to generate high-quality images without the need for delicate adversarial training. However, the current sampling process in DPMs is prone to violent shaking. In this paper, we present a novel reverse sampler for DPMs inspired by the widely-used Adam optimizer. Our proposed sampler can be readily applied to a pre-trained diffusion model, utilizing momentum mechanisms and adaptive updating to smooth the reverse sampling process and ensure stable generation, resulting in outputs of enhanced quality. By implicitly reusing update directions from early steps, our proposed sampler achieves a better balance between high-level semantics and low-level details. Additionally, this sampler is flexible and can be easily integrated into pre-trained DPMs regardless of the sampler used during training. Our experimental results on multiple benchmarks demonstrate that our proposed reverse sampler yields remarkable improvements over different baselines. We will make the source code available.

翻译：扩散概率模型（DPMs）已被证明无需精细的对抗性训练即可生成高质量图像。然而，当前DPM中的采样过程容易产生剧烈振荡。本文受广泛使用的Adam优化器启发，提出了一种新颖的DPM反向采样器。所提出的采样器可便捷地应用于预训练扩散模型，通过利用动量机制与自适应更新来平滑反向采样过程并确保稳定生成，从而获得更高质量的生成结果。通过隐式复用早期步骤中的更新方向，本采样器在高层次语义与低层次细节之间实现了更好的平衡。此外，该采样器具有灵活性，能轻松集成至预训练DPM中，且不受训练时所采用采样器的限制。在多个基准测试上的实验结果表明，我们提出的反向采样器在不同基线模型上均取得了显著提升。相关源代码将公开提供。